EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AO-RMM: A Scalable Riemannian Optimization Framework for Joint IRS Phase and Precoding Design in mmWave Systems

Mian Muhammad Kamal, Syed Zain Ul Abideen, Ijaz Khan, Basem Abu Zneid, Aadel Mohammed Alatwi

IEEE Open Journal of the Communications Society · 2026 · 7 · 4795-4806

DOI: 10.1109/ojcoms.2026.3690190 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Open Journal of the Communications Society, 2026. Verfasst von Mian Muhammad Kamal und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Intelligent reflecting surfaces (IRS) represent a promising paradigm for augmenting millimetre-wave (mmWave) multi-user communication systems through highly advanced control of the radio propagation environment. Nevertheless, simultaneously optimizing IRS phase shifts and transmitting precoding matrices remains computationally intensive due to the non-convexity of the objective and the unit-modulus constraints imposed on the IRS elements. To overcome these limitations without increasing the linear computational complexity, this paper proposes an alternating optimization Riemannian majorization-minimization (AO-RMM) methodology. The AO-RMM algorithm exploits the geometric properties of the complex unit circle to solve IRS phase constraints, and, using a majorization-minimization scheme, restates the non-convex problem in question as a sequence of convex subproblems. In comparison to the traditional semidefinite programming (SDP) approaches, which experience poor scalability and large computation cost, the proposed solution is linear in terms of the number of units to deploy per population, that is, it can be applied in large-scale deployments. Extensive simulation studies demonstrate that the AO-RMM framework provides a significant increase in aggregate throughput, even compared to randomly configured IRS setups, across a wide range of system parameters. The algorithm exhibits strong convergence behaviour, typically requiring 50 to 100 iterations, and remains effective even under discrete phase quantization. Scalability tests confirm that the throughput gains persist and even increase for IRS panels with up to 256 elements, validating the framework’s applicability to next-generation wireless architectures.

Zitieren

APA

Kamal, M. M., Abideen, S. Z. U., Khan, I., Zneid, B. A., & Alatwi, A. M. (2026). AO-RMM: A Scalable Riemannian Optimization Framework for Joint IRS Phase and Precoding Design in mmWave Systems. IEEE Open Journal of the Communications Society, 7, 4795-4806. https://doi.org/10.1109/ojcoms.2026.3690190

Vancouver

Kamal MM, Abideen SZU, Khan I, Zneid BA, Alatwi AM. AO-RMM: A Scalable Riemannian Optimization Framework for Joint IRS Phase and Precoding Design in mmWave Systems. IEEE Open Journal of the Communications Society. 2026;7:4795-4806. doi:10.1109/ojcoms.2026.3690190

BibTeX
@article{Kamal2026,
  author = {Kamal, Mian Muhammad and Abideen, Syed Zain Ul and Khan, Ijaz and Zneid, Basem Abu and Alatwi, Aadel Mohammed},
  title = {AO-RMM: A Scalable Riemannian Optimization Framework for Joint IRS Phase and Precoding Design in mmWave Systems},
  journal = {IEEE Open Journal of the Communications Society},
  year = {2026},
  volume = {7},
  pages = {4795--4806},
  doi = {10.1109/ojcoms.2026.3690190}
}
Kostenlos starten →